Structural Sensor Beamforming with Virtual Arrays for Noise Rejection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing acoustic beamforming technologies struggle to achieve high directivity and effective noise reduction in noisy environments, particularly for applications like smart speakers, due to limitations in sensor configurations and signal processing methods.
Innovation Solution
The use of structural vibration or strain sensing elements affixed to an elastic base surface, combined with frequency-domain weight application and singular value decomposition, allows for enhanced beamforming by simulating multiple sensor responses from a single sensor, achieving improved directional sensitivity and off-axis attenuation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple structural sensors are used to achieve high directivity and effective noise reduction, then the beamforming performance is improved, but the hardware complexity and cost increase
Solution Approach 1:
The patent uses virtual sensors to create copies of sensor responses through signal processing. By simulating the responses that would be obtained from multiple physical sensors using mathematical models and singular value decomposition, the system achieves the beamforming performance of multiple sensors while using only a single physical sensor, thereby reducing hardware complexity while maintaining measurement precision
Solution Approach 2:
The patent replaces the mechanical system of multiple physical sensors with a signal processing system. Instead of using multiple structural sensors mounted on the surface, the invention uses a single sensor combined with computational techniques (singular value decomposition, virtual sensor modeling) to achieve the same directional sensitivity and noise reduction, substituting mechanical complexity with computational processing
2Device complexity
If a single sensor is used to reduce hardware requirements, then device complexity is reduced, but the ability to achieve high directivity and noise reduction is limited
Solution Approach 1:
The patent transitions from the physical dimension to the computational dimension. Instead of adding more physical sensors in space, the invention creates additional sensor responses in the computational domain through virtual sensor modeling. This allows a single physical sensor to provide the directional sensitivity equivalent of multiple sensors by processing signals in the frequency and spatial domains
Solution Approach 2:
The patent changes the parameters of the single sensor system by applying frequency-domain processing and adaptive filtering. By modifying the signal parameters (frequency responses, phase relationships) through computational techniques, the system achieves enhanced directional sensitivity and noise reduction capabilities that would otherwise require multiple physical sensors
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach significantly enhances the intelligibility of target speech signals in noisy conditions, reducing word error rates by up to 191.8% and achieving up to 20 dB off-axis attenuation, while reducing hardware requirements.
Implementation Method 1
bending vibrations of the elastic base surface are excited by incoming acoustic waves
Implementation Method 2
Bending vibrations of the elastic base surface are excited by incoming acoustic waves
Data Source
AI summary
A method and system of detecting sound by acoustic beamforming with structural sensors wherein an array of sound sensors placed on an elastic base surface can pick up vibrations caused by incoming acoustic waves and acoustic beamforming enables highly sensitive directional listening. In some embodiments, only a single sensor is used affixed to an elastic base surface and the remaining sound reception may be constructed using virtual sensors that are created by extrapolation from the sound received by the single sensor.